Modelling Extortion Rackets Corinna Elsenbroich CRESS, University - - PowerPoint PPT Presentation

Modelling Extortion Rackets

Corinna Elsenbroich CRESS, University of Surrey

CSS, Lucca, September 2014

Collective Reasoning Moral Point of View

• 1. Team Reasoning

2.GLObal Dynamics of Extortion Racket Systems 3.An ABM of Team Reasoning in Extortion Rackets 4.Conclusions

The Next 10 Min

• 1. Team Reasoning

Game Theory

3,3 5,0 0,5 1,1

Prisoner’s Dilemma Basic Assumptions:

R’s Choice C’s Choice

1,1

C D D C

• Individuals are rational.
• Individuals are rational if they maximise their utility.
• Individuals assume the other player(s) are rational.
• What is rational for one agent to do depends on what is rational for another agent.
• Equilibria are solution to games.

A choice

The collective diagonal The individual lines 3,3 5,0 0,5 1,1 3,3 5,0 0,5 1,1

T eam Reasoning

Bacharach: Team reasoning is triggered by strongly interdependent games. A game is strongly interdependent if a Nash equilibrium is dominated by a Pareto

• ptimal solution. Team reasoning “just happens” when

the game is of the right kind, but it is conditional. Sugden: Mutually assured team reasoning

• 1. (1) I am a member of S.
• 2. (2) I identify with S and acknowledge U as its objective.
• 3. (3) In S, there is reciprocal reason to believe that each member of S identifies with S and acknowledges

U as the objective of S.

• 4. (4) In S, there is reciprocal reason to believe that each member of S endorses and acts on mutually

assured team reasoning.

• 5. (5) In S, there is common reason to believe that A uniquely maximizes U.

I should choose my component of A.

Rationality is reinterpreted as seeing the collective diagonal in a strongly interdependent game. Rationality is a choice to be made given a social group situation.

• Underlying dynamics of extortion rackets
• Societal conditions for success of extortion rackets
• Database of extortion cases in Sicily (Palermo)
• Simulation of Extortion Racket Systems (Rome)
• Qualitative data analysis of court files (Koblenz)
• Theoretical work, scenario building and testing (Surrey)
• Stakeholderboard (International Organised Crime Specialists)

GLODERS

Extortion and Extortion Rackets

Extortion is the extraction of money/favours from a victim using (the threat of) violence.

• An extortion racket is a systematic and continuous extortion of several victims.

The best thing for each individual is to pay and thus not to be punished but for everyone around them not to pay (and take the risk of punishment) draining the Mafia of income. The best thing for the group is to not pay the pizzo. . .

Extortion as a Collective Dilemma

This makes it a public goods game.

The best thing for each individual is to pay and thus not to be punished but for everyone around them not to pay (and take the risk of punishment), draining the Mafia of income. The best thing for the group is to not pay the pizzo. . .

Extortion as a Collective Dilemma

• 3. ABM of Extortion Racket Systems

13 p a y e x t

• r

t , p u n i s h

Extorters Entrepreneurs

An Existing Model

14 p a y e x t

• r

t , p u n i s h

Extorters Entrepreneurs

An Existing Model

Individualist Model

15 p a y e x t

• r

t , p u n i s h

Extorters Entrepreneurs Collective (e.g. Addio Pizzo)

understand

• neself as

a member stop paying

conceptual change

Team Reasoning in Extortion Rackets

• Income, Pizzo, Punishment Cost, Damages (kept constant)
• Extorter Radius
• Neighbourhood Radius
• Group Radius
• Number of Entrepreneurs
• Ratio of Collectivists and Individualists
• Number of Extorters

Team Reasoning in Extortion Rackets

• Extorters
• Entrepreneurs: Collectivists or Individualists

16

If punishment-probability * possible-damage < group-wealth + group-pizzo refuse. If punishment-probability * possible-damage < pizzo refuse.

• Income, Pizzo, Punishment Cost, Damages (kept constant)
• Extorter Radius
• Neighbourhood Radius
• Group Radius
• Number of Entrepreneurs
• Ratio of Collectivists and Individualists
• Number of Extorters

Team Reasoning in Extortion Rackets

• Extorters
• Entrepreneurs: Collectivists or Individualists

17

If punishment-probability * possible-damage < group-wealth + group-pizzo refuse If punishment-probability * possible-damage < pizzo refuse.

Analysis

18

0% Collectivists 50% Collectivists 100% Collectivists Overall

Low resistance, low punishment Increased resistance, low punishment High resistance, low punishment

Extorter Radius

Increases lead to resistance increase and punishment increase Increases lead to resistance and punishment increases but less pronounced Increases increases resistance only for high neighbourhood radii

Neighbourhood Radius

Increases lead to resistance decrease and punishment decrease Similar but for low extorter radii increases lead to increased resistance Increases lead to resistance increase

Group Radius

n/a Increases lead to increased resistance Increases lead to increased resistance

Scenario: Bottom-Up and Top-Down

• Bottom-up: Grass Roots Movement of Addio Pizzo

Entrepreneurs joining together in no longer paying the pizzo About a 10% point reduction in pizzo payments since 2004 In Model: Increasing numbers of Collectivists

• Top-down: Italian State fighting the Mafia

Wealth confiscated Mafioso imprisoned In Model: Decreasing extorter radius (reduced step by step) Mafiosi can no longer punish (taken out one by one)

Typical Runs - Resistance Levels

EX 5 NR 5 EX 5 NR 10 EX 10 NR 5 Ex 10 NR 10 GR 5 GR 10 GR 15 GR 20

Typical Runs - Bottom-Up and Top-Down

0% 50% 100%

Extorter Radius Constant (10) Neighbourhood Radius 10, Group Radius 10

0% 50% 100%

Extorter Radius Constant (10) Extorter Radius Decreasing (20-1)

Typical Runs - Bottom-Up and Top-Down

Neighbourhood Radius 10, Group Radius 10

0% 50% 100%

Extorter Radius Constant (10) Extorter Radius Decreasing (20-1) Decreasing number of active Mafiosi (1/200 steps)

Typical Runs - Bottom-Up and Top-Down

Neighbourhood Radius 10, Group Radius 10

• Conceptualising ERS as collective dilemmas can give a societal explanation
• Applying Team Reasoning shows that collective solutions are possible
• Scenario: Collectives increase resistance
• Resistance increases become particularly pronounced when collective action is combined

with judicial measures

• A decreasing radius curbing extorter territory
• Increasing the Mafiosi on strict surveillance so they can extort but no longer punish
• Conclusions and Future Work
• Generalise the model to implement Team Reasoning

as a genuine decision mechanism (Collective Reasoning as a Moral Point of View)

Conclusions Future Work

Thanks!